Devanagari Alphabet to Speech Conversion: A Research Perspective

Abstract

Text-to-speech (TTS) systems are instrumental in bridging the gap between textual content and auditory communication, especially for users with visual or reading disabilities. While such systems have matured for English and other Latin-script languages, they remain underdeveloped for Indian languages, particularly those written in the Devanagari script. This paper presents a detailed methodology for converting Devanagari alphabets into speech by addressing linguistic, phonetic, and computational challenges. A combination of rule-based and neural techniques for grapheme-to-phoneme mapping and speech synthesis is explored, aimed at enhancing clarity, accuracy, and naturalness of speech. The system is useful for applications in assistive technology, public information systems, language learning, and AI-based interfaces in Indian languages.